suyz526 commented on issue #13955: adding the gluon implementation of deformable convolution URL: https://github.com/apache/incubator-mxnet/pull/13955#issuecomment-456808132 The output channels of the convolution for the offset is hardcoded because its size is described in the paper. The infer_shape_partial() for the symbol.contrib.DeformableConvolution returns error: `mxnet.base.MXNetError: Error in operator deformable_conv0_deformableconvolution0: [10:48:09] src/operator/contrib/./deformable_convolution-inl.h:414: Check failed: oshape[2] == offsetshape[2] (1 vs. 0) output height must equal to offset map height ` `symbol.contrib.DeformableConvolution(data, offset, ...)` I have set the height and weight of 'data' and 'offset' to 0. However, the output of this function have the height and weight of 1. I don't understand why.
---------------------------------------------------------------- This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
